Tile and Wave Quantization in GPU Matrix Multiply
A matrix multiply is cut into thread-block tiles, and the tiles run in waves across the SMs. Calculate both rounding losses and read the steps in a throughput sweep.
9 min readConcept
Explore machine learning concepts related to Wave Quantization. Clear explanations and practical insights.
A matrix multiply is cut into thread-block tiles, and the tiles run in waves across the SMs. Calculate both rounding losses and read the steps in a throughput sweep.